File size: 1,669 Bytes
42a15d2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 |
import gradio as gr
import nltk
import pandas as pd
nltk.download('punkt')
from fincat_utils import extract_context_words
from fincat_utils import bert_embedding_extract
import pickle
lr_clf = pickle.load(open("lr_clf_FiNCAT.pickle",'rb'))
def score_fincat(txt):
li = []
highlight = []
for word in txt.split():
if any(char.isdigit() for char in word):
if word[-1] in ['.', ',', ';', ":", "-", "!", "?", ")", '"', "'"]:
word = word[:-1]
st = txt.index(word)
ed = st + len(word)
x = {'paragraph' : txt, 'offset_start':st, 'offset_end':ed}
context_text = extract_context_words(x)
features = bert_embedding_extract(context_text, word)
prediction = lr_clf.predict(features.reshape(1, 768))
prediction_probability = '{:.4f}'.format(round(lr_clf.predict_proba(features.reshape(1, 768))[:,1][0], 4))
highlight.append((word, ' In-claim' if prediction==1 else 'Out-of-Claim'))
li.append([word,' In-claim' if prediction==1 else 'Out-of-Claim', prediction_probability])
else:
highlight.append((word, ' '))
headers = ['numeral', 'prediction', 'probability']
dff = pd.DataFrame(li)
dff.columns = headers
return highlight, dff
iface = gr.Interface(fn=score_fincat, inputs=gr.inputs.Textbox(lines=5, placeholder="Enter Financial Text here..."), title="FiNCAT-2",description="Financial Numeral Claim Analysis Tool (Enhanced)", outputs=["highlight", "dataframe"], allow_flagging="never", examples=["In the year 2021, the markets were bullish. We expect to boost our sales by 80% this year.", "Last year our profit was $2.2M. This year it will increase to $3M"])
iface.launch()
|